CharlesLi
/

OpenELM-1_1B-DPO-full-max-reward-most-similar

Text Generation

alignment-handbook

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

OpenELM-1_1B-DPO-full-max-reward-most-similar / generation_config.json

CharlesLi's picture

Model save

3cad9f4 verified about 1 month ago

111 Bytes

	{
	"_from_model_config": true,
	"bos_token_id": 1,
	"eos_token_id": 2,
	"transformers_version": "4.45.1"
	}